Providing on-street parking navigation options is an area of interest for many navigation service providers and original equipment manufacturers (OEMs). To help provide such options, service providers often collect data on parking events (e.g., park-in or park-out events) to estimate parking availability information. However, data on parking events traditionally has been difficult to obtain because such data is often dependent on having an extensive network of installed parking sensors or a large of number of probe vehicles reporting parking events. Accordingly, service providers face significant technical challenges to collecting parking event data in a way that minimizes resource burdens associated with fixed parking sensors or maintaining probe vehicle networks.